A distributed database can be flexible, allowing routing of database changes and/or user message(s) between an arbitrary number of databases. Additionally, changes can be routed uni-directionally and bi-directionally, and allow user code to be run in multiple locations. However, configuration of a distributed database can be a time-consuming and frustrating experience. For example, users can encounter configuration issues in which a particular configuration of a distributed database does not behave as expected (e.g., without an error being raised).